Chapter 12 Laparoscopic Liver Resection for HCC: A European Perspective Luca Viganò and Daniel Cherqui Keywords Laparoscopic liver resection · HCC · Hepatic resection Hepatocellular carcinoma (HCC), the most common primary liver cancer, occurs in >90% of the cases on an underlying hepatic disease [1]. Screening programs allow diagnosis at an early stage where curative treatments can be proposed. These include liver resection, percutaneous radiofrequency ablation, and liver transplantation [1, 2]. Even if liver transplantation is the best treatment for early HCC by removing both the tumor and the underlying liver disease, shortage of donor organs and dropout from the waiting list limit its efficacy [3]. In recent years liver resection in cirrhotic patients became safer [4, 5] and achieved a key role in HCC treatment: in advanced tumors it is the only therapeutic option, while in early tumor it can be proposed as an alternative or a bridge to liver transplantation [6–9]. The vast majority of hepatic resections for HCC are stand-alone procedures, without any need for reconstruction, which should make them good candidates for a laparoscopic approach. However, diffusion of laparoscopic liver resection is still limited and few centers worldwide regularly perform it [10]. The reasons for the limited development of such an approach to date are threefold. First, technical prob- lems are anticipated and, indeed, the elementary maneuvers of open hepatic surgery (including manual palpation, organ mobilization, vascular control, and parenchymal transection) are thought to be difficult to reproduce laparoscopically. Second, there are anticipated hazards: hemorrhage may be more difficult to control laparoscopi- cally, especially in cirrhotic livers, and the risk of gas embolism may be increased by the use of pneumoperitoneum. The third problem is a fear of oncological inad- equacy and tumor spread. Although still limited in number of cases, publications about laparoscopic liver resection have increased in recent years and HCC has been one of the most common indications. In this chapter, we will review the various D. Cherqui (B) Department of Surgery, New York-Presbyterian/Weill Cornell, New York, NY, USA 185 K.M. McMasters, J N. Vauthey (eds.), Hepatocellular Carcinoma, DOI 10.1007/978-1-60327-522-4_12, C Springer Science+Business Media, LLC 2011 186 L. Viganò and D. Cherqui aspects of laparoscopic liver resection for HCC, including technical features, short- and long-term results. We will also briefly discuss the impact of laparoscopic liver resection on the treatment strategy of HCC. Feasibility: Technique and Indications In comparison with open hepatectomy series, the number of published papers about laparoscopic liver surgery is very low [10]. At present only 14 studies (includ- ing 1 multicentric) reported 50 or more cases [11–24] (Table 12.1). Interestingly the majority of them have been published in the last 2 years [11, 13, 18–24]. An increasing proportion of malignant diseases have been treated and HCC was the most common indication. The feasibility of laparoscopic liver resection has been the main criterion studied to date. Despite the increasing number of reported series, in expert centers laparo- scopic approach ranges from 5 to 30% [11, 13, 15, 16, 22] and only some recent series reported higher rates, reaching 50–80% [18, 20, 21]. On our part, over the past 12 years (1996–2008), we performed 174 laparoscopic liver resections out of 782 hepatectomies (22.3%) [25]. Considering HCC, the proportion of laparoscopic resection was higher, about 30% (69 of 229) and reached 39.4% in the last 4 years of our experience [25]. Table 12.1 Series of laparoscopic liver resections including more than 50 cases Author Year # Proportion of LLR on total LR Malignant lesions HCC Descottes [14] a 2003 87 NR 0% 0% Mala [15] 2005 53 44% 89% (47) 2% (1) Kaneko [16] 2005 52 17% NR 77% (40) Vibert [12] 2006 89 NR 73% (65) 18% (16) Cai [17] 2006 62 NR 32% (20) 29% (18) Dagher [11] 2007 70 15% 54% (38) 34% (24) Koffron [18] b 2007 273 NR 37% (103) NR Chen [19] 2008 116 NR 100% (116) 100% (116) Topal [13] 2008 109 28% 71% (77) NR Buell [20] 2008 253 NR 42% (106) 14% (36) Cho [21] 2008 128 NR 61% (78) 45% (57) Sasaki [23] 2008 82 29% 93% (78) 45% (37) Inagaki [24] 2009 68 NR 76% (52) 43% (36) Cherqui [25] 2009 174 22% 63% (110) 40% (69) a multicentric study b only pure laparoscopic and hand-assisted laparoscopic hepatectomies included NR: data not reported; LLR: laparoscopic liver resection; LR: liver resection 12 Laparoscopic Liver Resection for HCC 187 Surgical Technique State-of-the-art equipment is required. The use of two monitors is recommended. Although some groups use 0 ◦ laparoscopes [11, 12], 30 ◦ laparoscopes are preferred by most authors. Patient Positioning We suggest two different positions according to lesion site. For lesions located in segments 2 through 5 (the majority of cases), the patient is placed in the supine position, with lower limbs apart (Fig. 12.1). The surgeon stands between the legs with one assistant on each side. For patients with lesions of segment 6 scheduled for atypical resection or segmentectomy, the left lateral decubitus position may be used in order to expose the lateral and posterior aspect of the right liver (Fig. 12.2). In this case the surgeon is on the ventral side of the patient. In case of laparoscopic right hepatectomy, supine position with lower limbs apart is preferred. Some authors prefer supine position with the surgeon stand on patient side and the assistant on the opposite one [18]. Pneumoperitoneum A problem concerning laparoscopic liver surgery is the pneumoperitoneum itself. The risk of gas embolism due to hepatic vein lesions during parenchymal Fig. 12.1 Port placement for resection of lesions located in segments 2–5 and for right hepatectomy. The patient is in supine position with lower limbs apart and the surgeon between the legs. Numbers shown represent trocar sizes in millimeters 188 L. Viganò and D. Cherqui Fig. 12.2 Port placement for resection of lesions located in segment 6. The patient is in left lateral decubitus for right lobe mobilization and posterior exposure. The table can be turned to the right to reapply the right lobe and gain anterior access. Numbers shown represent trocar sizes in millimeters transection has been suggested. Transesophageal echocardiography study in animal model demonstrated gas embolism in almost all animals undergoing laparoscopic liver resection with cardiac arrhythmia in two-thirds of cases [26]. In order to avoid it, gasless laparoscopy has been proposed [27]. However, gas embolism occurrence in clinical practice is extremely low [28]. In 2002, Biertho et al. reviewed published laparoscopic liver resections and reported only 2 cases of possible gas embolism over about 200 procedures [29]. In recent series [11, 18, 20, 30] and in our expe- rience [22], few cases of transient mild cardiovascular alteration due to embolism occurred without clinical consequences. Carbon dioxide pneumoperitoneum mini- mizes risk of gas embolism as compared to air and low pneumoperitoneum pressure further reduces its incidence [31]. Electronic monitoring of intra-abdominal pres- sure is required and should be maintained at less than 14 mm Hg. Gas embolism occurrence has been also related to argon beam coagulation which increases endo- abdominal pressure leading to increased risk of gas embolism [32]. To date CO 2 pneumoperitoneum is considered safe and gasless laparoscopy is no longer in use. Port Sites Positioning and Hand Assistance Positioning of port sites is different according to tumor site and it is shown in Figs. 12.1 and 12.2. Many variants have been described. The position of trocar for the laparoscope can be higher on the midline or more lateral on the right side in case of right liver resection [11, 12]. Hand-assisted laparoscopy is used by several authors [18, 33–35]. It consists in the placement through an 8-cm incision of a gas-tight port permitting the introduc- tion of a hand in the abdomen. The assisting hand allows tactile feedback while palpating the liver and it may help in abdominal exploration, mobilizing the liver, provides gentle retraction, and helps during parenchymal transection. In addition, 12 Laparoscopic Liver Resection for HCC 189 in case of bleeding hand compression allows easier hemostasis. For its proponents, this technique may render laparoscopic liver resection safer and more accessible. Koffron et al. recently proposed a wide use of hand assistance in order to increase the proportion of patients that can benefit from laparoscopic-assisted approach [18]. In our experience, hand assistance has been used in selected cases (about 10%) of right hepatectomies or limited resections of posterior right segments to facilitate when liver mobilization or parenchymal transection can be difficult. Pedicle Clamping Intermittent clamping (15-min clamping and 5-min release periods) can be per- formed whenever necessary. Our group demonstrated that in patients with normal cardiac function laparoscopic pedicle clamping is safe and well tolerated [36, 37]. However, it is used less often and the majority of recent resections have been performed without any clamping even in cirrhotic patients [25]. Liver Mobilization and Inflow/Outflow Control Several techniques have been described which cannot be detailed here. Our usual technique is briefly depicted. In left lateral sectionectomy, the round, falciform, and left triangular ligaments and the lesser omentum are divided. Dissection of the falciform ligament is con- tinued to the level of the inferior vena cava and the insertions of the hepatic veins. Parenchymal transection is carried out until the portal pedicles of segments 2 and 3 are exposed. The pedicles are then divided using linear staplers. Left hepatic vein is divided at the end of parenchymal transection by linear stapler [38]. In limited resections, parenchymal transection is carried out along decided tran- section lines. Portal pedicles and hepatic veins are controlled as they are encountered during transection. In limited right-sided resections, the right triangular ligament is divided, taking advantage of the lateral position of the patient. Parenchymal transection is then carried out. Laparoscopic right hepatectomy includes dorsal decubitus position, initial divi- sion of the right portal pedicle, right liver mobilization, taping of right hepatic vein if feasible and transection. Hand assistance can be used. Hand port is introduced through a right iliac or flank transverse incision. Surgeon’s left hand or assistant’s right hand helps mobilizing the liver and compresses in case of bleeding. Parenchymal Transection The main technical challenge of laparoscopic liver resection remains hemorrhage during parenchymal transection, especially in cirrhotic patients. Several devices have been developed with the aim to perform more bloodless and accurate parenchy- mal transection. These devices have not proved to be indispensable during open resections. However, in laparoscopic surgery, the simple principles of transection are more difficult to apply and some of the newly designed technologies are required. 190 L. Viganò and D. Cherqui The Ultrasonic Aspirator – The ultrasonic dissector selectively destroys liver parenchyma and spares vessels and bile ducts that can be selectively controlled. It does not have hemostatic properties. In our experience ultrasonic dissector is par- ticularly useful in deep parenchymal transection, especially in right hepatectomy, to selectively identify and control vessels and bile ducts. The Ultrasonic Scalpel – Also called harmonic scalpel, it has the major advan- tage to cut and coagulate at the same time. In the laparoscopic procedures, easy handling and rapid action are major advantages of this device. It can be particularly recommended for the superficial parts of transection (2 cm in depth). However, it is a blind instrument which should be used with caution when deeper liver parts are reached because of the risk of vascular injuries to larger vessels, especially to hepatic veins. The Vessel Sealing System – The vessel sealing system uses low-frequency bipo- lar current and seals vessels up to 7 mm in diameter. Its use is rather similar to that of the ultrasonic scalpel, and it includes a knife that cuts after sealing. Radiofrequency-Assisted Hepatic Resection (Habib Laparoscopic Sealer 4XL R ) – The radiofrequency probe inserted along transection line generates pre-coagulation. Subsequent cut along coagulated line can be performed. Many advantages have been suggested: easy and bloodless transection, mainly in atypical resections; it can be helpful in wedge resections, in which visualization of tran- section planes and bleeding control may be more difficult; induced necrosis may improve safe surgical margins. Further studies are needed to evaluate its role in laparoscopic liver surgery. Stapler Hepatectomy – Linear stapler devices are widely applied in laparoscopic liver surgery for portal pedicles and hepatic veins division. Recently, some authors proposed their use for parenchymal transection [39]. After the transection line is marked and the liver capsule is incised with diathermy, liver parenchyma can be divided with repeated applications of linear vascular staplers. According to its pro- ponents, this technique allows fast and safe resection. However, vascular and biliary injuries can occur during blind transection and this technique does not allow fine control of margins and requires that tumors are located remotely from the transec- tion line. In addition, the cost of this method is high and increases with the number of applications required. We have not favored this approach and further studies are necessary to clarify safety of stapler hepatectomy. Other Devices – Many other devices have been proposed, such as water jet dissection, microwave-based devices, curettage and aspiration device, and monopo- lar irrigated coagulation devices, but available data do not allow any conclusive evaluation. Specimen Extraction In all cases, the specimen is placed in a plastic bag and extracted through a sepa- rate incision, either along a previous appendectomy incision or a new supra-pubic horizontal incision. Enlarged port site can also be utilized. Fragmentation must of course be avoided to allow proper pathological evaluation. 12 Laparoscopic Liver Resection for HCC 191 Indications Indications to laparoscopic hepatectomy do not differ from those of open surgery. Technical feasibility has been reported as the only limiting factor [11, 12, 18, 20, 21, 40]. In order to select the best candidates for laparoscopic liver resection, two criteria have been considered by all authors. Tumor Location HCC located in antero-lateral segments of the liver (segments 2–6, so-called laparo- scopic segments, Fig. 12.3) and scheduled for wedges, segmentectomies, and left lateral sectionectomies are the best indications for laparoscopic approach [11, 12, 41]. Laparoscopic right hepatectomy can be planned for HCC located anywhere in the right lobe with the exception of those close to the hilum or the hepato-caval junc- tion, because of the risk of major vascular or biliary injury. The role of laparoscopy for lesions requiring resections of segments 7, 8, and 1 is not yet codified. Even if they have been traditionally considered non-laparoscopic segments because of diffi- cult visualization of surgical field, hand-assisted laparoscopy and thoracoscopy have been proposed in such location [18, 19, 23, 34, 42]. Cho et al. recently reported a series of 36 patients with lesions located in postero-superior segments (Sg7-8-4a-1) treated by pure laparoscopic approach [21, 43]. Even if 30% of cases underwent a right hepatectomy, anatomic segmentectomies, atypical resections, and right poste- rior sectionectomies have also been performed. Similarly, laparoscopic resections of segment 1 have been recently reported [13, 18, 20]. Further studies are necessary to confirm feasibility and reproducibility of these procedures. Fig. 12.3 The “laparoscopic segments.” Shaded areas are considered consistent with laparoscopic resection 192 L. Viganò and D. Cherqui Tumor Size Except for exophytic lesions which are easy to resect by laparoscopy, even if large in size, laparoscopy is usually not recommended for HCC exceeding 5 cm of diameter [11, 19, 21–23, 44, 45]. Even if some authors did not adopt this criterion [12, 18, 20], laparoscopic liver resection cannot be recommended for large intrahepatic lesions because of difficult tumor mobilization and risks of rupture or inadequate margin. Liver Function Liver function is an essential component of selection of patients considered for liver resection for HCC. Results of liver surgery in cirrhotic patients significantly improved thanks to a strict patient selection based on their liver function and future remnant liver volume [4, 5, 46]. In open surgery only Child–Pugh A patients with a future remnant liver over 40% are considered for liver resection. Presence of portal hypertension is not an absolute contraindication to liver surgery but indications have to be cautiously discussed on a case-by-case basis, and we would only consider a limited resection for such patients [3, 47, 48]. The same criteria should be adopted for laparoscopic liver resection. In case of peripheral nodules requiring atypical resections, some authors proposed laparoscopic liver resection in patients with poor liver function (Child–Pugh B) [24, 34, 49, 50]. Laparoscopic approach allows easy Fig. 12.4 Laparoscopic atypical resection of segment 2 as a “bridge” to liver transplantation for HCC in Child–Pugh B patient. (a) Preoperative MRI. (b)The specimen. (c) Postoperative CT scan, 1 month after the laparoscopic resection 12 Laparoscopic Liver Resection for HCC 193 resection of peripheral nodules preserving the abdominal wall and its collateral veins and reducing fluid dispersion. Good outcomes have been reported, but further data are required to codify these indications. In our center few Child–Pugh B patients have been treated (Fig. 12.4); indications are discussed on a case-by-case basis and retained in a limited number of cases. Evaluation of Laparoscopic Liver Resection No randomized study on the efficacy of laparoscopic liver resection has so far been published. Studies on laparoscopy in other areas of abdominal surgery may provide a worthwhile analogy. There are few randomized trials even for common opera- tions that compare open and laparoscopic approaches and even fewer that have demonstrated any superiority of laparoscopy; conversely, none has demonstrated any superiority of laparotomy. For example, randomized studies comparing laparoscopy and mini-laparotomy for cholecystectomy have failed to demonstrate the superiority of one approach over the other [51, 52]. However, few would dispute that laparoscopy is now the standard approach for elective cholecystectomy. This suggests that in the absence of a clear difference between laparotomy and laparoscopy and provided that the same result can be achieved, surgeons favor laparoscopy. Another analogy can be drawn from the COST randomized trial of open vs laparoscopic colectomy for colon cancer [53]. This study was designed as a non-inferiority trial and enrolled 872 patients. It demonstrated similarity for recurrence and survival and a slight advantage to laparoscopy for hospital stay and analgesic requirements (1 day reduc- tion for each item). The conclusion was that the laparoscopic approach is an acceptable alternative to open surgery for colon cancer. A randomized study of open vs laparoscopic liver resection is of course desirable, but it will be difficult to conduct because of the variability of the indications, types of resections, and types of techniques used. It will also require a large number of patients which will be difficult to accrue. At the present time only retrospective and case–control comparisons are available. An increasing number of papers reported outcomes of laparoscopic liver resec- tions. The majority of them are focused on short-term outcomes, such as hospital stay and analgesic requirements. These are important but vary according to local practice. Other outcome measures, such as re-operations, incisional hernias, and bowel obstructions, are potential advantages which remain to be demonstrated. Finally, there are very few oncological results available. Short-Term Outcomes Few papers have been specifically addressed on HCC. We will consider our own experience and the literature on laparoscopic liver resection presently available focusing on HCC or including HCC cases. 194 L. Viganò and D. Cherqui Table 12.2 Short-term results of laparoscopic liver resection in patients affected by HCC (series including more than 10 patients) Author Year # Major hep Mortality Morbidity Conversion Shimada [45] 2001 17 0% 0% 6% (1) 0% Teramoto [42] a 2003 11 0% 0% 18% (2) 1 converted case excluded Kaneko [57] 2005 30 0% 0% 10% (3) 3% (1) Tang [30] 2006 17 0% 0% NR NR Vibert [12] 2006 16 NR 6% (1) NR NR Belli [61] 2007 23 0% 4% (1) 22% (5) 4% (1) Cai [62] 2008 24 NR 0% 0% NR Dagher [58] 2008 32 13% (4) 3% (1) 25% (8) 9% (3) Cho [21] 2008 57 18% (10) 0% 16% (9) 5% (3) Chen [19] 2008 116 3% (4) 0% 6% (7) 5% (6) Buell [20] 2008 36 9.7% (3/31) b 9.7% (3/31) b 29% (9/31) b NR Sasaki [23] 2008 37 0% 0% 3% (1) 0% Santambrogio [59] 2009 19 0% 0% 11% (2) 1 converted case excluded Huang [35] 2009 27 NR 0% 19% (5) NR Lai [60] 2009 25 4% (1) 0% 16% (4) 4% (1) Inagaki [24] 2009 36 0% 0% NR NR Cherqui [25] 2009 69 7% (5) 0% 22% (15) 13% (9) a Laparoscopic and thoracoscopic resections; b Data detailed only for cirrhotic patients (31/36 HCC) NR: data not reported In cirrhotic patients, liver resections, even minor ones, carry a high risk of complications, including ascites, jaundice, and encephalopathy [54, 55]. Specific benefits from laparoscopic approach have been suggested: it might offer the advan- tage of preserving the abdominal wall and its collateral veins resulting in less portal hypertension, less need for fluids, and improved re-absorption of ascites [40]. Short-term results of series including more than 10 consecutive laparoscopic liver resections for HCC are detailed in Table 12.2. Across more than 2000 published laparoscopic liver resections, 10 postoper- ative deaths have been reported (less than 0.5%) [11–13, 20, 21, 44, 56]. Six out of ten deaths occurred in cirrhotic patients [11, 12, 20, 44]. Buell et al. [20] recently reported more than 250 laparoscopic liver resections and observed a sig- nificantly increased mortality in cirrhotic patients in comparison with non-cirrhotic ones (9.7% vs 0.3%). These data underline that liver surgery in cirrhotic patients has to be considered at increased risk, even if laparoscopically performed. In the literature, morbidity rates after laparoscopic liver surgery ranged from 5 to 20% [11, 13, 14–23]. Considering cirrhotic patients they tended to be higher, about 10–30% [19, 21, 25, 35, 42, 45, 57–62]. In the above-mentioned series of Buell et al., morbidity was 29% in cirrhotics vs 14% in non-cirrhotics (p = 0.02) [20]. Morino et al. observed increased hospital stay and blood loss in case of cirrhosis [63]. In . safe and gasless laparoscopy is no longer in use. Port Sites Positioning and Hand Assistance Positioning of port sites is different according to tumor site and it is shown in Figs. 12.1 and 12.2 vein if feasible and transection. Hand assistance can be used. Hand port is introduced through a right iliac or flank transverse incision. Surgeon’s left hand or assistant’s right hand helps mobilizing. technique allows fast and safe resection. However, vascular and biliary injuries can occur during blind transection and this technique does not allow fine control of margins and requires that tumors